Child Care Center Feasibility Study

A count in every year since FY1991

The child care record has no blank years: every row from FY1991 carries a count in both programs. The first year shows 216 7(a) loans for $46.2 million and 34 504 loans for $7.3 million; the full record holds 24,935 7(a) loans for $12.71 billion and 5,496 504 loans for $3.72 billion. The FY2017 to FY2026 rows that the rest of this page analyses contain 8,128 of those 7(a) loans and $6.75 billion of the dollars, and 1,959 of the 504 loans with $2.02 billion, so the recent window holds most of the class's dollars on a minority of its loans. When too many of a year's loans lack a usable industry code in SBA's files, the source note makes that year's count a floor, and any early row so marked should be read as a minimum. For a study, the length of the record is useful in itself: a lender can see the class across several lending cycles, and a study that claims a market is ready for another center should show the center carrying its debt in the weak years of that record as well as the strong ones.

A first peak in FY2005, then the FY2009 fall

In the early record 7(a) loans were many and small. Approvals rose from 216 in FY1991 to 757 in FY1995, yet those 757 loans carried only $117.7 million, less than the $124.5 million lent on 598 loans in FY1996. The count reached its first peak at 1,092 in FY2005 for $285.3 million and held at 1,040 in FY2007, the year 504 reached 284 loans and $157.1 million, a 504 count no later year has matched. Then both programs contracted: 7(a) to 835 loans in FY2008 and 508 in FY2009, 504 to 249 and then 173. The relevance for a study is the exit. If an owner has to sell or refinance a center in a year like FY2009, the pool of buyers able to finance a purchase through these programs shrinks with the approvals, and a lender holding a long loan will want to see that the center's value does not depend on a buyer arriving in a strong year.

After FY2010 the loans grew

Recovery after FY2009 came first in dollars. FY2011 brought 576 7(a) loans for $333.3 million, more money than FY2005 lent on 1,092 loans, and FY2014 reached $400.0 million on 619 loans. By FY2016, 794 loans carried $640.7 million. Set the two highest count years side by side: 1,092 7(a) loans in FY2005 carried $285.3 million, and 1,145 in FY2025 carried $832.0 million. A child care 7(a) loan today is a much larger credit than the one in the early rows, and a study that calls a request typical has to measure it against the FY2017 to FY2026 loans, not against the class's history as a whole, or the lender will find the comparison flattering and discount it.

FY2019 to FY2026: a low, a record and larger tickets

The window this page analyses has its own cycle. 7(a) approvals fell from 741 in FY2019 to 590 in FY2020, came back to 794 in FY2021, slipped to 637 in FY2022 and then climbed through 772 and 975 to 1,145 in FY2025. FY2026 brought 869 loans and $811.3 million, close to FY2025's $832.0 million on far fewer loans, so the latest year's loans ran larger than those of the record year before it. A large request therefore meets plenty of recent precedent, and the lender will compare it with what it approved in FY2025 and FY2026. A smaller request will be judged more on the operating numbers than on any building, which is where the study's enrollment, tuition and staffing work carries the file.

504 keeps its own rhythm

504 lending to the class does not track 7(a). In FY2022, the year 7(a) dropped to 637 loans, 504 rose to 270 loans and $274.2 million, its highest count since FY2007. 504 dollars then reached $288.6 million in FY2025, the program's largest year for child care centers, and stood at $244.7 million on 183 loans in FY2026. In no year of the record did 504 lend more to the class than 7(a); even in FY2025 the comparison was $288.6 million against $832.0 million. For a center that will own its building, the recent 504 rows give the study a solid base of comparable financings, and a lender will look to the study for the reason 504, with its larger median loan, suits the project better than a 7(a) loan would.

Houston and Dallas-Fort Worth outweigh most states

Houston-Pasadena-The Woodlands has 419 loans for $609.4 million at an $803,000 median, more dollars than any state except Texas, Florida and California; Georgia's statewide total is $599.1 million. Dallas-Fort Worth-Arlington adds 404 loans for $591.5 million at $924,650, and with Austin-Round Rock-San Marcos (139 loans, $189.2 million) and San Antonio-New Braunfels (67, $91.0 million) most of the Texas record of 1,151 loans sits in these metros. Both large Texas metros sit above the state median of $762,000. Houston is also among the few metros with a printed charge-off floor above 0.00%: 4.0%, higher than the Texas floor of 2.9%. A Houston study will be read against large loans and a floor above the state's, and the lender will want the competitive survey to cover a market where its own record shows heavy child care lending.

New York-Newark and the small-loan Northeast

New York-Newark-Jersey City leads the metro table with 654 loans and $490.5 million at a $322,000 median and a 5.2% charge-off floor. That median sits above New York State's $150,000, partly because the metro spans New Jersey, whose state median is $452,150, and partly because upstate loans run small: Rochester's 44 loans carry $13.7 million at $152,000 and Buffalo-Cheektowaga's 25 carry $11.5 million at $100,000. New England reads the same way. Boston-Cambridge-Newton has 256 loans for $155.0 million at $250,000 and a 5.5% floor, Worcester 35 loans at $248,400, Providence-Warwick 76 at $223,500 with a 0.00% floor, and Portland-South Portland 42 at $194,000. Massachusetts prints 8.3%, above the Boston floor, so the losses in the state's record are not only a Boston story. Lenders in these markets are used to small child care loans; a large request needs a study that shows why this center is a different credit from the typical loan in their files.

Large-ticket metros in the Southeast and the West

The largest medians sit in a different set of metros. Charleston-North Charleston's 30 loans carry $50.9 million at a $1.2 million median, the highest in the table; Atlanta-Sandy Springs-Roswell has 404 loans for $533.4 million at $973,700, above Georgia's own $882,500; Raleigh-Cary has 79 loans at $908,800, Denver-Aurora-Centennial 151 at $879,800, Richmond 44 at $874,000, Seattle-Tacoma-Bellevue 154 at $802,100 and Charlotte-Concord-Gastonia 127 at $757,000. Atlanta carries most of Georgia's record, 404 of its 486 loans, and Denver most of Colorado's, 151 of 202. In these metros a lender's comparable files are full of large loans, and the study for a new center has to support a land and building budget at that scale with cost evidence, rents or sale prices for comparable buildings, and a fill schedule that covers debt from the opening months.

Inside a state, metros diverge

Ohio's low state median of $187,650 resolves into very different metros: Columbus has 203 loans for $138.1 million at $303,000, Cleveland 103 at $185,000, Akron 40 at $100,000 and Cincinnati 79 at $100,000. Wisconsin shows the same split: Milwaukee-Waukesha's 102 loans carry $55.3 million at $255,000, while Madison's 61 carry $52.8 million at $540,000. Pennsylvania has Pittsburgh at 62 loans and $290,800 and Harrisburg-Carlisle at 29 loans and $739,500. San Jose-Sunnyvale-Santa Clara and San Diego-Chula Vista-Carlsbad have almost the same count, 77 and 78, but $91.3 million against $48.6 million. A lender reading a study in any of these states will compare the request with its metro, so the study should cite the metro median and loan count rather than the state's, and explain any gap between the project and that metro.

Florida and California spread their loans

Florida's record spreads across many metros: Miami-Fort Lauderdale-West Palm Beach with 266 loans and $235.4 million, Tampa-St. Petersburg-Clearwater with 171 and $185.5 million, Orlando-Kissimmee-Sanford with 154 and $170.4 million, Jacksonville with 78, Palm Bay-Melbourne-Titusville with 46, Port St. Lucie with 34, North Port-Bradenton-Sarasota with 29 and Lakeland-Winter Haven with 27. Medians run from $457,000 in Lakeland to $748,000 in Orlando, and Tampa and Jacksonville print 0.00% floors. California's metros also differ on loan size: Los Angeles-Long Beach-Anaheim leads with 232 loans at $510,000, San Francisco-Oakland-Fremont has 177 at $793,000, Riverside-San Bernardino-Ontario 61 at $446,000 and Sacramento-Roseville-Folsom 55 at $482,400. For a Florida or California study the statewide numbers mix markets that do not resemble each other, and the lender will expect the metro row to anchor the argument.

Withheld floors

Most metro charge-off floors are withheld, including those of Chicago-Naperville-Elgin with 422 loans, Atlanta and Dallas-Fort Worth, and Cincinnati has fewer than 30 loans in its cohort. Withholding hides a value; it does not signal a small one, and this page never prints a number in its place. Minneapolis-St. Paul-Bloomington, Tampa, Detroit-Warren-Dearborn, Milwaukee-Waukesha, Jacksonville and Providence-Warwick print 0.00%. Where the metro row is silent, the study should use the state floor and the class floors of 2.7% for 7(a) and 0.14% for 504, and say that it has. A lender in Chicago, where the metro floor is withheld, will read the study against the Illinois floor of 1.8% and the class floor, and a study that names those numbers and builds its downside case on them answers the question before the lender asks it.

Where the quartiles fall

7(a) child care loans run from a lower quartile of $112,500 through the $387,000 median to an upper quartile of $950,000; 504 loans run from $343,500 through $759,000 to $1.5 million. The 504 lower quartile sits below the 7(a) median, so the programs overlap across a wide middle range, but their tops differ: a quarter of 504 loans exceed $1.5 million, while a 7(a) loan above $950,000 is already in the largest quarter of its program. A lender places a request on this scale before it reads anything else. A 7(a) request near $112,500 is small enough that the lender will ask what it finances and whether the center already operates; one above $950,000 will be compared with the largest and most closely studied loans in the class.

7(a): many small loans, most dollars at the top

Counting the size bands in the table's order, the 7(a) record holds two different books inside one program. The bottom band, loans of $150,000 or less, is the most crowded: 2,584 loans, yet only $184.6 million. The fifth band, the largest any child care 7(a) loan reached, holds 954 loans and $3.41 billion, more loans than the fourth band's 937 and far more dollars than any other band; no 7(a) loan sits above it. The third band carries 2,296 loans and $1.45 billion, the second 1,357 and $360.6 million. Because the class shares its industry code with in-home care, part of the smallest band may finance a provider's home rather than a center. For a fifth-band loan the lender expects a complete market, cost and operating study, read against 954 precedents; a bottom-band loan calls for a narrower study that says plainly what the money buys and why the center's cash flow covers it.

504 sits in the middle and upper bands

504 loans sit higher. The smallest band holds 104 loans, with its dollars withheld; the second 403 loans and $103.8 million; the third 670 loans and $421.7 million, the most of any 504 band; the fourth 498 loans and $730.0 million; and the fifth 282 loans and $737.3 million, the largest 504 dollar band by a narrow margin. Only 2 loans under 504 exceed the fifth band, and their dollars are suppressed. A 504 center loan is therefore usually a third, fourth or fifth band credit, and the study has to support real estate value at that scale: land, construction or purchase cost, the classroom and play-space build-out the license requires, and an appraisal that can stand if the center changes hands.

At the top, 7(a) outnumbers 504

At the largest sizes the class leans on 7(a), not 504. The fifth band holds 954 7(a) loans for $3.41 billion against 282 504 loans for $737.3 million, and the fourth band 937 7(a) loans against 498. A borrower might expect 504 to carry the biggest center projects; in this record most large child care loans were written under 7(a). For the study the consequence is an allocation question. A large 7(a) loan may finance a building, an operating business with its enrollment, or both, and the lender will want the study to separate the value that rests on the property from the value that rests on the center's enrollment and staff, since the second depends on the center continuing to operate.

Cancellation is common in both programs; losses are mostly 7(a)

Approvals fail to fund at nearly the same rate in both programs, 12.2% for 7(a) and 12.5% for 504, both floors counted on FY2017 to FY2024 approvals. Losses differ sharply: the 7(a) charge-off floor is 2.7% against 0.14% for 504. The five-year column adds a timing point. For 504 the floor within 5 years and the floor to date are the same, 0.14%; for 7(a) the five-year floor is 1.5% against 2.7% to date, which indicates that part of the 7(a) losses in the FY2017 to FY2021 disbursement vintage came after the fifth year. A study for a 7(a) loan should not end its downside case when the enrollment ramp ends; the lender will want to see coverage hold through later risks such as a new competitor nearby, a change in subsidy rates or the loss of key staff.